1 PROJECT SUMMARY 2 The spike trains of retinal ganglion cells (RGCs) are the sole source of visual information to the brain. In mice, 3 approximately 40 RGC types send information to 46 subcortical targets. Each RGC type encodes a distinct set 4 of features of the visual scene (e.g., global luminance, local contrast, color, the orientation of edges, and 5 specific patterns of motion). Much attention has focused on dissecting the retinal circuits that give rise to the 6 diverse RGC responses. By comparison, which RGC types provide input to specific subcortical targets, how 7 subcortical circuits process retinal information, and how different projections from subcortical targets regulate 8 diverse behaviors remains mostly unknown. This proposal addresses these questions, focusing on the 9 ventrolateral geniculate nucleus (vLGN) of the thalamus. The vLGN is the least understood part of the lateral 10 geniculate complex, which consists of the dorsolateral geniculate nucleus, the intergeniculate leaflet, and the 11 vLGN. In Aim 1, we will combine retrograde tracing with two-photon guided patch-clamp recordings and 12 quantitative single-cell reconstructions to identify all RGC types that provide input to vLGN and to characterize 13 the visual information they convey. Most neurons in vLGN are inhibitory, and, as a population, vLGN neurons 14 project to diverse midbrain nuclei. We hypothesize that vLGN parses diverse retinal inputs into target-specific 15 output signals that mediate feature-selective gain control in downstream circuits. Our preliminary data reveal 16 that different vLGN neurons project to the olivary pretectal nucleus (OPN) and the lateral posterior (LP) 17 nucleus of the thalamus. The OPN mediates pupillary light responses, which contribute to light adaptation in 18 the visual system. The OPN receives excitatory input from the retina and inhibitory input from vLGN. Whereas 19 the function of the retinal input to vLGN has been studied in detail, how inhibition from vLGN shapes light 20 responses of OPN neurons and regulates pupillary light responses is unknown. In Aim 2, we will record from 21 OPN-projecting vLGN neurons in awake mice to characterize their light responses. We will then silence OPN- 22 projecting vLGN neurons to determine their influence on OPN neurons and pupillary light responses. The LP is 23 involved in midbrain pathways that mediate innate defensive responses to threatening visual stimuli (i.e., 24 looming stimuli). The LP receives excitatory input from superior colliculus and inhibitory input from vLGN. The 25 excitatory input from superior colliculus is required for normal looming responses. The function of the inhibitory 26 input from vLGN is unknown. In Aim 3, we will record from LP-projecting vLGN neurons in awake mice. We will 27 then silence LP-projecting vLGN neurons to elucidate their influence on LP neurons and on innate defensive 28 responses to looming stimuli.